Quiz #9

Question 1		True / False	(1 points)
	Question:	With programmed I/O the processor issues a command to an I/O module and then busy-waits for the operation to be be completed before proceeding.
		True False

Question 2 True / False (1 points)

Question: With direct memory access (DMA) I/O, the I/O device does not interrupt the CPU.

True
False

Question 3 True / False (1 points)

Question: With interrupt-driven I/O, the CPU interrupts the I/O device when it is ready for data.

True
False

Question 4 True / False (1 points)

Question: Direct memory access (DMA) I/O can "steal" memory cycles from the CPU, which can cause the CPU to run slower than if there were no DMA I/O going on.

True
False

Question 5 Multiple Choice (1 points)

Question: On a moveable head disk system, the amount of time it takes to position the head at the track is called the _______.

rotational delay
access time
data transfer time
track selection delay
head positioning delay
seek time
none of the above

Question 6 Multiple Choice (1 points)

Question: On a moveable head disk system, the amount of time it takes for the beginning of the sector to reach the head, once the head is positioned on the track, is called the ________.

rotational delay
access time
data transfer time
head positioning delay
track selection delay
none of the above

Question 7 Multiple Answer (1 points)

Question: Which of the following generally takes the shortest time?

data transfer from disk drive, after head is in position
moving head to the proper track
waiting for disk to rotate the proper sector under the head
defragmenting a filesystem
checking the consistency of a filesystem

Question 8 Multiple Choice (1 points)

Question: Suppose a series of disk I/O requests has been made, for tracks 125, 25, 150, and 50 (in that order), and the head is currently serving a request at track 60, having moved there from track 10. What is the order the four pending requests would be served by the SCAN algorithm?

125, 25, 150, 50
50, 25, 125, 150
125, 150, 50, 25
125, 150, 25, 50
25, 50, 125, 150
150, 125, 50, 25

Question 9 Multiple Choice (11 points)

Question: Suppose a series of disk I/O requests has been made, for tracks 125, 25, 150, and 50 (in that order), and the head is currently serving a request at track 60, having moved there from track 10. What is the order the four pending requests would be served by the C-SCAN algorithm?

125, 25, 150, 50
50, 25, 125, 150
125, 150, 50, 25
125, 150, 25, 50
25, 50, 125, 150
150, 125, 50, 25

Question 10 Multiple Choice (1 points)

Question: Suppose a series of disk I/O requests has been made, for tracks 125, 25, 150, and 50 (in that order), and the head is currently serving a request at track 60, having moved there from track 10. What is the order the four pending requests would be served by the SSTF algorithm?

150, 125, 50, 25
50, 25, 125, 150
125, 150, 50, 25
125, 150, 25, 50
25, 50, 125, 150
150, 125, 50, 25

Question 11 True / False (1 points)

Question: The SSTF disk scheduling policy minimized the average seek time over a number of arm movements.

True
False

Question 12 True / False (1 points)

Question: The N-step-SCAN and the FSCAN disk scheduling algorithms both group requests into batches, and bound the variation in wait times by applying FIFO ordering between batches.

True
False

Question 13 True / False (1 points)

Question: Several forms of RAID provide improvements in data throughput rate, between disk and memory.

True
False

Question 14 True / False (1 points)

Question: Several of the forms of RAID can tolerate failure (loss) of an entire single disk drive without loss of data.

True
False

Question 15 Matching (1 points)

Question: Match each of the types of RAID with the descriptive phrase that fits it best.

Match Question Items

1. RAID 0

2. RAID 1

3. RAID 2

4. RAID 3

5. RAID 4

6. RAID 5

7. RAID 6

Answer Items

A. can tolerate two disk failures

B. block-interleaved parity, single disk bottleneck

C. redundant, doubles number of disk drives

D. non-redundant, only gains performance

E. redundant via Hamming code, single-disk bottleneck

F. redundant via bit-interleaved parity, distributed parity

G. block-interleaved parity, distributed parity

Question 16 Fill in the Blank (1 points)

Question: What is the value of the expression "(A xor B) xor A",
where A is the bit string 1010 and B is the bit string 1100?

Question 17 Multiple Choice (1 points)

Question: Why would the LRU replacement algorithm be more practical for disk cache management than for virtual memory page replacement?

disk accesses exhibit more locality of reference
memory is always getting cheaper
there is more time between disk I/O requests than between page references
there is more time between page references than between disk I/O requests
none of the above; it is not more practical

Question 18 Multiple Choice (1 points)

Question: Which of the following is true of disk caching?

the content of disk blocks is hashed to find the block in the cache
the device number and logical block number are hashed to find the block in the cache
the buffer address is hashed to find the block in the cache
the search time to find out whether a block is in the cache is O(n)
the search time to find out whether a block is in cache is O(log n)
the search time to find out whether a block is in cache is O(sqrt(n))

Question 19 True / False (1 points)

Question: For disk cache management, LRU is the most commonly used policy.

True
False

Question 20 True / False (1 points)

Question: With a disk cache, improved performance can be achieved by writing out batches of sectors that are located close together on the disk (and freeing the cache slots), rather than waiting to write out a sector when a cache slot is actually needed.

True
False

Question 21 True / False (1 points)

Question: LRU always outperforms frequency-based replacement policies, for disk caching.

True
False

Question 22 Multiple Choice (1 points)

Question: Which of the following is most similar in function to a disk cache, to the point that the OS is likely to combine the implementations into a single subsystem?

translation lookaside buffer
memory cache
file system
database system
virtual memory

Question 2		True / False	(1 points)
	Question:	With direct memory access (DMA) I/O, the I/O device does not interrupt the CPU.
		True False

Question 3		True / False	(1 points)
	Question:	With interrupt-driven I/O, the CPU interrupts the I/O device when it is ready for data.
		True False

Question 4		True / False	(1 points)
	Question:	Direct memory access (DMA) I/O can "steal" memory cycles from the CPU, which can cause the CPU to run slower than if there were no DMA I/O going on.
		True False

Question 5		Multiple Choice	(1 points)
	Question:	On a moveable head disk system, the amount of time it takes to position the head at the track is called the _______.
		rotational delay access time data transfer time track selection delay head positioning delay seek time none of the above

Question 6		Multiple Choice	(1 points)
	Question:	On a moveable head disk system, the amount of time it takes for the beginning of the sector to reach the head, once the head is positioned on the track, is called the ________.
		rotational delay access time data transfer time head positioning delay track selection delay none of the above

Question 7		Multiple Answer	(1 points)
	Question:	Which of the following generally takes the shortest time?
		data transfer from disk drive, after head is in position moving head to the proper track waiting for disk to rotate the proper sector under the head defragmenting a filesystem checking the consistency of a filesystem

Question 8		Multiple Choice	(1 points)
	Question:	Suppose a series of disk I/O requests has been made, for tracks 125, 25, 150, and 50 (in that order), and the head is currently serving a request at track 60, having moved there from track 10. What is the order the four pending requests would be served by the SCAN algorithm?
		125, 25, 150, 50 50, 25, 125, 150 125, 150, 50, 25 125, 150, 25, 50 25, 50, 125, 150 150, 125, 50, 25

Question 9		Multiple Choice	(11 points)
	Question:	Suppose a series of disk I/O requests has been made, for tracks 125, 25, 150, and 50 (in that order), and the head is currently serving a request at track 60, having moved there from track 10. What is the order the four pending requests would be served by the C-SCAN algorithm?
		125, 25, 150, 50 50, 25, 125, 150 125, 150, 50, 25 125, 150, 25, 50 25, 50, 125, 150 150, 125, 50, 25

Question 10		Multiple Choice	(1 points)
	Question:	Suppose a series of disk I/O requests has been made, for tracks 125, 25, 150, and 50 (in that order), and the head is currently serving a request at track 60, having moved there from track 10. What is the order the four pending requests would be served by the SSTF algorithm?
		150, 125, 50, 25 50, 25, 125, 150 125, 150, 50, 25 125, 150, 25, 50 25, 50, 125, 150 150, 125, 50, 25

Question 11		True / False	(1 points)
	Question:	The SSTF disk scheduling policy minimized the average seek time over a number of arm movements.
		True False

Question 12		True / False	(1 points)
	Question:	The N-step-SCAN and the FSCAN disk scheduling algorithms both group requests into batches, and bound the variation in wait times by applying FIFO ordering between batches.
		True False

Question 13		True / False	(1 points)
	Question:	Several forms of RAID provide improvements in data throughput rate, between disk and memory.
		True False

Question 14		True / False	(1 points)
	Question:	Several of the forms of RAID can tolerate failure (loss) of an entire single disk drive without loss of data.
		True False

Question 16		Fill in the Blank	(1 points)
	Question:	What is the value of the expression "(A xor B) xor A", where A is the bit string 1010 and B is the bit string 1100?

Question 17		Multiple Choice	(1 points)
	Question:	Why would the LRU replacement algorithm be more practical for disk cache management than for virtual memory page replacement?
		disk accesses exhibit more locality of reference memory is always getting cheaper there is more time between disk I/O requests than between page references there is more time between page references than between disk I/O requests none of the above; it is not more practical

Question 18		Multiple Choice	(1 points)
	Question:	Which of the following is true of disk caching?
		the content of disk blocks is hashed to find the block in the cache the device number and logical block number are hashed to find the block in the cache the buffer address is hashed to find the block in the cache the search time to find out whether a block is in the cache is O(n) the search time to find out whether a block is in cache is O(log n) the search time to find out whether a block is in cache is O(sqrt(n))

Question 19		True / False	(1 points)
	Question:	For disk cache management, LRU is the most commonly used policy.
		True False

Question 20		True / False	(1 points)
	Question:	With a disk cache, improved performance can be achieved by writing out batches of sectors that are located close together on the disk (and freeing the cache slots), rather than waiting to write out a sector when a cache slot is actually needed.
		True False

Question 21		True / False	(1 points)
	Question:	LRU always outperforms frequency-based replacement policies, for disk caching.
		True False

Question 22		Multiple Choice	(1 points)
	Question:	Which of the following is most similar in function to a disk cache, to the point that the OS is likely to combine the implementations into a single subsystem?
		translation lookaside buffer memory cache file system database system virtual memory